Comparative evaluation of performance measures for shading correction in time-lapse fluorescence microscopy

Area of Science:

• Cell biology
• Microscopy imaging
• Image processing

Background:

• Time-lapse fluorescence microscopy is crucial for cell biology research.
• Shading artefacts, caused by uneven illumination or camera issues, lead to inaccurate single-cell intensity measurements.
• Existing shading correction methods and performance evaluation metrics lack standardized application for real-world data.

Purpose of the Study:

• To review state-of-the-art shading correction and performance evaluation methods.
• To objectively compare 10 popular shading correction techniques on artificial and real datasets.
• To identify the most suitable performance measure for evaluating shading correction in time-lapse fluorescence images.

Main Methods:

• Literature review of shading correction and performance evaluation techniques.
• Implementation and application of 10 shading correction methods on diverse datasets (2 artificial, 4 real).
• Quantitative performance assessment using multiple metrics, with a focus on the coefficient of joint variation (CJV).

Main Results:

• The coefficient of joint variation (CJV) was identified as the most applicable performance measure for time-lapse fluorescence images.
• A novel shading correction method was developed, outperforming established techniques on tested real-world data.
• Extensive validation confirmed the superiority of the proposed method and the utility of CJV for evaluation.

Conclusions:

• The coefficient of joint variation (CJV) is a robust metric for evaluating shading correction in time-lapse fluorescence microscopy.
• The newly proposed shading correction method offers improved accuracy and performance on real biological imaging data.
• This work provides a framework for objective comparison and development of shading correction techniques in cell biology.